Leukemic Stem Cell Expression Signatures Identify Novel Therapeutics for Acute Myeloid Leukemia
Bibliographic record
Abstract
Abstract Acute myeloid leukemia (AML) has a relatively poor 5-year survival rate of 30% in adults, largely due to high rates of relapse, which is thought to be driven by chemoresistant leukemic stem cells (LSCs). A therapy would need to eradicate all LSCs to obtain durable responses in AML patients. Unlike typical cancerous cells, LSCs are often quiescent, allowing them to evade standard therapies and serve as a reservoir for relapse. Thus, the identification of novel anti-LSC therapies could result in the improvement of the clinical outcome of patients suffering from AML. To identify compounds predicted to impede LSC function while not affecting normal hematopoietic stem cells (HSCs), we probed datasets of drug-gene interaction with our previously defined LSC and HSC gene expression signatures generated from 16 primary AML samples and 4 pooled cord blood samples (Eppert et al. Nat Med, 2011), and an additional LSC-signature from 86 AML patients (unpublished data; Ng et al. Nature, 2016). We identified 152 molecules predicted to target LSCs without harming HSCs. Eighty-three of these molecules were screened against a primary AML sample (8227) with a known LSC-containing population (confirmed as CD34+CD38- by xenotransplantation). Viability and phenotype were assessed by flow cytometry. We identified multiple hits in two classes of drugs: 3 drugs in a subclass of steroids and 3 drugs in a specific group of ion pump inhibitors, that preferentially eliminated CD34+CD38- cells over CD15+ blast cells in 8227. Many of these candidates are already in use for other clinical settings. The steroids induced differentiation in the lower nanomolar range (IC50: 0.44-1.30nM), shown by a depletion of primitive CD34+CD38- cells and subsequent expansion of terminally differentiated CD15+ cells. The ion pump inhibitors preferentially targeted the CD34+CD38- cells (IC50 CD34+CD38-: 11.55nM, 24.69nM, 21.78nM) over the CD15+ blast cells (IC50 CD15+: 17.72nM, 32.73nM, 45.09nM). To determine the toxicity of our candidates towards normal stem and progenitor blood cells, we exposed CD34+ enriched human cord blood to the steroids and ion pump inhibitors. All candidates had minimal effects against the CD34+ cord blood cells compared to CD34+ 8227 cells. We then tested the compounds against multiple primary AML samples and observed that at low nanomolar concentrations (0.3-4.0nM) the steroids differentiated AML #9642 (M4e, NPM1-WT, FLT3-WT) but had no effect on AML #9706 (M1, NPM1-WT, FLT3-ITD). For the ion pump inhibitors, AML #184 (M4, NPM1-MT, FLT3-TKD) and AML #9642 were sensitive while AML #116 (M0, NPM1-WT, FLT3-WT) and AML #9706 were more resistant. This variability in sensitivity to both groups of candidates suggests a possible link to subtype or genotype to responsiveness. To evaluate the effect of the steroids and ion pump inhibitors on functional leukemic progenitor cells, 8227 cells were treated with the steroids, ion pump inhibitors or control (DMSO) and then assessed by colony formation unit assay. Treatment with the steroids or ion pump inhibitors resulted in a 2-fold decrease in progenitors (colony formation) compared to DMSO. Next, we will expose a panel of AML patient samples to the candidate compounds in vitro to further elucidate whether there is a link between certain clinical or genotypic features and responsiveness. We will also perform xenotransplantation of treated AML samples into immunodeficient mice to assess the effects on functional LSCs. Overall, we have shown that both the subclass of steroids and specific ion pump inhibitors identified by our bioinformatic approach have anti-AML and anti-LSC properties against particular AML samples in vitro . Future results from our study will provide valuable insight to LSC biology and can lead to new therapeutic approaches for targeting the LSCs, the root of therapy resistance and relapse in AML. Disclosures No relevant conflicts of interest to declare.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".